General Electric Co. v. Nitro Tungsten Lamp Co.

266 F. 994 | 2d Cir. | 1920

HOUGH, Circuit Judge

(after stating the facts as above). [1] That Langmuir's lamp is in a sense a new thing is not denied; nor is it questioned that the thing has proved fit for use, especially for display and advertising purposes, and in that field achieved a measure of success'. But the path of invention in respect of electric lighting has now been trodden by the feet of so many men of talent, and research even into its bypaths been so thorough, that the first defense offered is that “the teachings of the prior art are such as to leave no room for the exercise of patentable invention on the part of Dr. Langmuir.”

As a second defense it is urged that the claims in suit are (a) too indefinite to be regarded as legal definitions of invention, and (b) are the result of proceedings in the Patent Office of such a nature as to deny the claims a construction rendering defendant an infringer. Irt one sense the case deals with a concrete article of commerce; in another, it is plain that this decision may be thought to affect the rights of the public in respect of lighting devices quite different from anything as yet constructed, so far as this record shows, under the protection of this patent. The field of experimental investigation has been, we think, greatly enlarged by the practical success pf Langmuir’s tungsten nitrogen lamp, and merely noting the admitted fact that the patent may possibly hereafter be construed to cover filaments other than tungsten, and gaseous fillings not named in the specification, this decision is to be restricted (as was done in the court below) to ascertaining whether the particular claims in suit confine to plaintiff the right of making, etc., the alleged infringing article.¹

The subject, then, of this litigation, is what defendant has made and sold, viz. the “comet” or “standard multiple gas-filled tungsten” lamp, having (in a typical lamp) a tungsten filament .003 inch in diameter, helically coiled to a coil diameter of about .017 inch; dry nitrogen in the bulb at a pressure (cold) of about 700 mm.; and a starting efficiency of .82 to .95 watt per candle. After 500 hours’ operation at rated voltage, the temperature of the filament will still be approximately 2520° C., or about 400° higher than that of one of same metal in a vacuum lamp operating at about 1 watt per candle. This is so close a copy of plaintiff’s lamp in the 150-watt form, and every claim before us so plainly reads upon it, that it states the case to ask whether the man who first made this thing — admittedly new to commerce when plaintiff offered it to the world — was an inventor.

One basis for the answer to this question is the amount of knowledge imputed by law to that necessary legal fiction, the “man skilled in the art,” when Langmuir applied for his patent in 1913. We may be brief, for there is little controversy over the historical facts; the inferences therefrom produce the conflict. The early history of incandescent electric lighting as revealed by evidence has been written by Wallace and Lacombe, JJ., in the Edison Lamp Case, 52 Fed. 300, 3 C. C. A. 83; affirming (C. C.) 47 Fed. 454. In that litigation Mr. Edison (testifying in May, 1890) said:

“The discovery 1 made was that a line filament of carbon under the conditions í liad did not disintegrate to any extent. That was the discovery as set forth in my patent, but ® * * it required invention to carry out the discovery which I made.”

The decisions referred to scarcely do more than judicially validate the patentee’s estimate of his own performance. Edison’s lamp, perhaps somewhat aided by success iu litigation, established for many years as the fundamentals of incandescent electric lighting the “conditions” to which Edison referred, viz. a filament, a vacuum, and an all-glass inclosing globe.

Vacuum meant a commercial vacuum as established by experience and experiment, and the Edison combination was not defeated by the introduction into the bulb of a small quantity of bromine gas for the purpose of improving “the stability of the carbon and the diminution of tlie blackening of the glass of the lamp.” Whether the bromine was serviceable or not is immaterial, for its presence, as long as a commercial vacuum was maintained, neither affected the courts (Edison, etc., Co. v. Waring, etc., Co. [C. C.] 59 Fed. 358, affirmed 69 Fed. 645, 15 C. C. A. 700), nor led to essential modification in the art. By various subsidiary inventions, such as the “gelter” devices (Malignani v. Jasper Marsh, etc., Co. [C. C.] 180 Fed. 442), Edison’s original lamp, with a life of 600 hours at about 6% watts per candle, was improved as time went on; but the filament remained carbon, and that filament still wore out, and in wearing blackened the bulb. The improvements did not wholly fulfill the belief or hope expressed by Mr. Edison in his 1890 testimony in saying:

“I 1 bought that perhaps, having- gotten rid of all oxygen, this disintegration would not be so large a factor as to prevent the use of a lamp for commercial purposes, and the discovery I made was that this did not take place under.the conditions of a high, stable vacuum. Q. What causes the blackening of the globe in an ordinary incandescent lamp? A. I have spent over a hundred thousand dollars in trying to find out, and I don’t know.”

Improvement in vacuum having apparently ■reached its ultimate, search for a new and improved filament to put into the vacuum resulted in the discovery by Just and Hanaman of the tungsten filament (patent No. 1,018,502), broadly upheld in General, etc., Co. v. Laco-Philips Co., 233 Fed. 96, 147 C. C. A. 166, to which may be added the improvement of Coolidge in respect of preparing or treating tungsten (patent No. 963,872).

Thus in 1913 the tungsten vacuum lamp (dating from about 1905) with a life oí a thousand hours at scarcely over 1 watt per candle, with the filament heated to about 2,100° C., represented the last word in incandescent electric lighting; and it seems to us that the situation as practical men viewed it was summed up, when in that year Mr. Alexander Siemens publicly said that it was—

“very doubtful whether it will'be possible to construct a much more economical glow lamp (than the vacuum tungsten), so that the consumer will have to look for further economy to the improvement and cheapening of the electric supply.”

The foregoing .summary is the view of the practical man of skill, but there were other phenomena known to the scientist. In any scheme of incandescent lighting, the energy supplied to the filament is subject to three forms of expenditure; it radiates in light waves; it escapes as heat, and is dissipated by the movement or “sweeping action” of whatever gaseous material may be in contact with the incandescent substance. These forms of loss are duly set forth in the patent as radiation, conduction, and convection. The object of the vacuum was to do away with convection by dispensing with -the gaseous surrounding. Yet it had long been appreciated that raising the heat of the incandescent light giver produced more light, which was, of course, desirable; but, the higher the heat, the more rapid the deterioration of the incandescent filament. The reasons for this deterioration were obscure. The “air washing” of Edison (duly carried into the opinion in the Edison Lamp Case) had become an obsolete phrase. But, passing other theories, we think it uncontradicted on this record that Dr. Langmuir was the first to establish (so far at least as tungsten is concerned) that to heighten heat hastened a true evaporation of the heated substance. It was also known that the introduction of an inert gas, usually nitrogen, info the lamp bulb, delayed what we may now call the evaporation of the filament; but it brought back and accentuated convection losses, which the vacuum had minimized with world-famous commercial results.

Every design or machine is in a sense a compromise, and it is unquestionably true that during the whole history of electric incandescent lighting there had been suggestions that it would be possible to utilize a gaseous filling (usually nitrogen) in the bulb that the life of the filament might be extended. Thus Prof. Anthony said before the American Institute of Electrical Engineers in 1894:

“Witli such a gas [bromine] in the lamp and a properly proportioned filament, the initial efficiency may be carried as high as in the vacuum lamp, and the efficiency and illuminating power will be well preserved to the end.”

Yet he qualified the assertion by saying:

“If we could 'in any way * * * further] cheek the circulation of the gas within the lamp chamber, we should still further check the waste of the filament and prolong the life of the lamp.”

But neither he nor his principals ever did it.² When the rare and refractory metals, thorium et al., began to be hopefully regarded as possible supplanters of carbon filaments, it was suggested (as in Sander, British 14,411 of 1901) that if the lamp bulb be filled with “a mixture of nitrogen and carburetted hydrogen the lamp will also have an excellent lighting capacity.” But it was never reduced to practice. This history of hopeful suggestion is sufficiently elaborated in the opinion of the District Court in this case. 261 Fed. 606.

To us it seems fairly certain that both the commercial and the theoretic art had been put on the wrong road by Mr. Edison through the disclosure of his patent No. 274,295, when, in 1883, he proposed to fill a carbon filament bulb with an “inert gas,” viz. nitrogen, at a pressure oí about two-thirds of an atmosphere, and stated as one of the means of his hoped-for success that—

“The filament before carbonization may be reduced to a smaller cross-section iban usual heretofore in order to produce reduction of radiating surface.”

This was unquestionably wrong, and we fail to find any suggestion in the evidence regarding scientific discussion and theoretic statement that, before Rangmuir, any scientist disclosed to the world that in the nitrogen-filled bulb the loss by convection would be diminished, and a working compromise reached between that loss and the gain in filament life, by increasing the “effective size” of the filament itself.

The patentee on this record was also the' first to establish another crucial and illuminating fact, viz. that with a tungsten filament nitrogen of ordinary or commercial dryness was worse than useless. There is no evidence that any one before Rangmuir had even suggested that, in order to prevent a bulb-blackening fatal to the success of any lamp, the removal of water vapor, not only from the gaseous filling, but from the bulbs and filaments themselves, must be carried to an extent theretofore unused, if not undreamed of. In short, it was not through isolated experiments, but by correlating deductions from many of them, that the patentee produced in 1913 a lamp which in the larger sizes can operate on the usual 115-volt current with an efficiency as low as a half a watt per candle.

This thing defendant has copied, and of it defendant’s expert says:

What “made possible the success of the gas-fllled lamp was tlie fact that the drawn wire tungsten filament was developed. The general design of a gas-fllled lamp was well known in the art at that time (1. e., 1913). It was only by substituting tungsten for the materials which were not satisfactory that a satisfactory gas-filled lamp was developed.”

W'e are compelled to think this the result of partisanship. It is not true that merely putting nitrogen in a tungsten vacuum lamp will make anything worth having; it is true that a nitrogen-filled bulb will permit the “overrunning” of even a carbon lamp for a little while, and for its short, but expensive, life the lamp will be brilliant. But not even as a laboratory experiment had a gas-filled lamp lived a commercial life before Rangmuir. This court had no difficulty in Marconi, etc., Co. v. De Forest, etc., Co., 243 Fed. at page 564, 156 C. C. A. 258, in finding invention in the transference of a “laboratory problem to a new and very practical field of usefulness.” The difference between that case and this is that not even the laboratory problem had been solved before this patentee did it, and we hold with the lower court that invention was present.

[2] The defense resting on the wording and history of the claim may be thus presented:

When the application was filed, the solicitor suggested (inter alia) the following comprehensive claim:

“Tlte combination of a lamp bulb, a filling therein of dry nitrogen at a relatively high pressure, and a filament of tungsten.”

It may be said generally that from claims of this breadth or simplicity the patentee was compelled to recede and accept (inter alia) the claims before us. A claim is both a definition and an assertion. The definition must be justified by the disclosure; the assertion stands alone. The claim as proposed would have found support in the disclosure, but as an assertion it said too much, in that it would have read on some of the experimental failures of previous scientists, if only any known tungsten filament were substituted for carbon. But such a lamp would not have worked at all, and it would not have been Langmuir’s contribution to.the sum of human knowledge.

He did not invent any nitrogen-filled bulb with any tungsten filament in it, but a special article of special proportions and a carefully stated co-ordination of parts. We'therefore think it plain that the claims in suit have been narrowed from those first propounded, and are not open to the familiar objections arising from abandonment under compulsion. Langmuir abandoned nothing that he is trying to get in this suit, and that is the commonest and best test of the rule’s applicability.

[3] Nor is it perceived' that any claim before us is open to the charge of indefiniteness. On the contrary, if such a claim as that originally suggested and above quoted had been allowed, it would surely have been accused of attempting “to corral the art by the use of comprehensive indefinite terms,” and to “foreclose broadly against the future.” Kintner v. Atlantic, etc., Co. (D. C.) 249 Fed. at page 77. We reassert the statement of Eibel, etc., Co. v. Remington, etc., Co., 234 Fed. at page 632, 148 C. C. A. 390, that it is well not to attempt definition, but to consider the alleged indefiniteness of a claim in the light of the facts of each particular case. On the present facts, it is clear that it was (1) impossible to define the parts of Langmuir’s lamp in millimeters or other dimensional units; and (2) no,such effort was necessary to instruct the skilled man how to make the lamp of the patent.

It was impossible to give exact measurements, because the economic object of the lamp was to diminish the wattage per candle, and dimensions must be proportioned to the designed wattage; i. e., substantially to the size of the lamp — something to be worked out according to rules presumably long familiar to a competent electrical engineer. It was unnecessary to do more than state the limits of invention in terms of result, because the results desired are not functional, and do indicate limits in terms of lamp life and candle power which are likewise presumably quite familiar to any competent electrician. When a claim defines achievement in words no broader than the disclosure, and in phrases which, as interpreted by competent workers in the art, tell one how to do what the patentee did, it can rarely be called indefinite.

For the reasons foregoing, and confining ourselves wholly to the claims in suit and defendant’s lamp, we are of opinion that the decision below was right, and it is affirmed, with costs.

1 Plaintiff apparently sued on all the patent claims; the trial judge rested decision solely on those first enumerated; defendant only appealed. For this reason, also, we express no opinion as to any of the other claims.

2 Prof. Anthony was the expert for defendant in Edison, etc., Co. v. Waring, etc., Co., supra, and his Institute address is substantially his evidence in that ease.